Among the phonograph cylinders held at the Budapest Museum of Ethnography are over 3,000 with Bartók’s field recordings made before the outbreak of WW1 in the Romanian-speaking territories of what was then Hungary. The Bucharest Institute of Ethnography and Folklore has more than 10,000 further cylinders with field and studio recordings made in the 1930s in Romania by Constantin Brăiloiu and his collaborators. For Bartók the cylinders were, at the time, a means to an end: the detailed notation and classification of folk music in pursuance of his growing interest in ‘comparative musical folklore’. For Brăiloiu the chief interest lay in his conviction that oral traditions are underpinned by system no less than those of notated classical music. Today the two collections present us with an incomparably rich repository for further research, not least in understanding the psychology of how lyrical verses are stored in and retrieved from memory in notationless cultures. This is a topic that has resonance in medieval songs whose notation is similarly is a matter of record rather than prescription. The most teasing aspect of such songs has long been considered to lie in their rhythms. Unsurprisingly, perhaps, since it was not until the mid-thirteenth century that the utterly novel idea was rehearsed that certain kinds of music are precisely measurable. Surprising as it may seem, the cylinder recordings, taken in conjunction with Bartók’s writings on ‘parlando-rubato’ rhythm and Brăiloiu’s on ‘syllabic giusto’, have much to tell us about singing ‘beyond measure’ in medieval times.